domains / families / science / bionemo-kermt-embed

bionemo-kermt-embed

separate

Extract per-molecule embeddings from any encoder-bearing KERMT checkpoint. Use a local checkpoint or optionally download a pinned Hugging Face model bundle using HF_TOKEN if configured. Run containerized embedding extraction and write model bundles, per-readout .npy embeddings, canonical SMILES, and validity arrays to user-selected host directories.

When to use

• Extract per-molecule embeddings
• Use local KERMT checkpoint
• Download pinned Hugging Face model bundle
• Run containerized embedding extraction
• Write per-readout .npy embeddings
• Write canonical SMILES
• Write validity arrays

When not to use

• Checkpoints with no encoder

Skill page

Source